KR20230051361A - Search engine optimization by selective indexing - Google Patents

Abstract

An exemplary embodiment of the present invention relates to a system and method for identifying a useful view item page for search engine optimization. The system and method provide an improved point compared to an existing system that does not have a process of identifying or selecting a useful view item page to be used for driving traffic in a display site. The system and method disclosed in the present invention improve an existing system by predicting possibility of future traffic for a product provided based on a plurality of product level factors as an input variable, and by identifying selection of a view item page corresponding to a product having the possibility of the best future traffic for maximizing natural search traffic guided to a linked site of a corresponding view item page.

Description

Search engine optimization through selective indexing {SEARCH ENGINE OPTIMIZATION BY SELECTIVE INDEXING}

This application claims priority to U.S. Application Serial No. 15/131,139, filed on April 18, 2016, which claims priority to U.S. Provisional Application No. 62/187,016, filed on June 30, 2015, all of which The entirety is incorporated herein by reference.

Search engine indexing collects, analyzes and stores data for fast and accurate information retrieval. The purpose of storing indexes is to optimize speed and performance when finding relevant documents for a search query. Without an index, a search engine can search all documents in a corpus, which takes a lot of time and computing power. For example, an index of 10,000 documents may be queried in milliseconds, but sequentially scanning all the words in a large document of 10,000 may take several hours. The additional computer storage required to store the index, and the increased time required to perform updates offset the time saved during information retrieval.

Some embodiments are described by way of example and not limitation, in conjunction with the accompanying drawings.
1 is a network diagram illustrating a client-server system in which one exemplary embodiment may be deployed.
2 is a block diagram illustrating components of a selective indexing system suitable for selecting valuable view item pages for selective indexing, in accordance with some demonstrative embodiments.
3 is a flow diagram illustrating operation of the selective indexing system in identifying useful view item pages based on indexed items, in accordance with some demonstrative embodiments.
Figure 4 shows how the execution of a machine learning (ML) model depends on the decision threshold.
5 illustrates an exemplary form of a computer system capable of executing a set of instructions that cause the machine to perform any one or more of the methods described herein.

Reference will now be made in detail to specific embodiments for carrying out the subject matter of the present invention. An embodiment may be practiced without some or all of these details. It will be understood that the foregoing disclosure is not intended to limit the scope of the claims to the described embodiments. On the contrary, it is intended to cover alternatives, modifications and equivalents that may be included within the scope of this disclosure as defined by the appended claims. Also, well-known features may not have been described in detail in order to avoid unnecessarily obscuring the present invention.

Exemplary embodiments identify useful View Item Pages (VIPs) so that search engines (eg, Google™, BING™) can find sites linked to these VIPs (eg, linked sites, Systems and methods for improving (eg, increasing) traffic to EBAY™, etc.). This system and method provide significant improvements over existing systems that do not involve identifying or selecting useful VIPs to use to drive traffic at the display site. The systems and methods described herein predict the probability of future traffic for a given product based on a number of product-level factors as input variables, and to the corresponding VIP's linked site (e.g., EBAY). We improve on the previous system by identifying a selection of VIPs corresponding to products with the highest probability of future traffic to maximize guided natural search traffic.

The probability of future traffic for a given item is determined by building a machine learning (ML) model. ML models make data-driven predictions and decisions based on a set of inputs. In some demonstrative embodiments, ML models are created via gradient boosted machine learning techniques (GBM). GBM is a machine learning technique for regression and classification, which creates a predictive model in the form of an ensemble of weak predictive models (eg, a decision tree). ML models are trained and tested by providing sample data and target variables (eg organic search traffic values) as input data. The ML model can then build a correlation between the sample data and the target variable output.

An ML model calculates the probability of future traffic for a given product (or set of products) based on input variables, including item-level factors and previous search engine optimization (SEO) performance metrics for VIPs corresponding to a product or set of products. do. Previous SEO performance metrics include, for example, organic search traffic, view count, listing type (e.g., good till canceled (GTC), auction format, Dutch auction, buy now, etc.), rank-weighted impression count (rank weighted impressions count, meta category, watch count, predicted quality, average price of similar products, price, number of listings by seller of the product, time on site, seller feedback score, bounce count ) (e.g., the number of sessions in which a user left the site from an entry page without interacting with that page), previous bounce count, quantity of product, status, and quantity sold. In some embodiments, organic search traffic through VIP is used as a target variable by the ML model to calculate the probability of future traffic based on GBM techniques. Organic search traffic is the frequency with which a product is returned as an organic search result as opposed to being presented as an advertisement. Based on the organic search traffic calculated by the ML model, the system is configured to make an index or no-index decision for a product (or set of products).

If the probability of future traffic to a set of VIPs is calculated, and an index or no-index decision is made for a product (or set of products), then the ML model calculates the link (e.g., of the VIP) based on the index and no-index decisions. Identify one or more VIPs from a VIP set to maximize traffic to your site. For example, the system can access the VIP of a product classified as "index" and insert an HTML meta tag within the VIP indicating that the VIP is an index page. Meta tags are used by search engines to determine whether a page should be returned as a search result. Thus, web pages containing the "index" meta tag (e.g., product listing pages, VIPs) can be searched and included within the search result set, while pages containing the "no index" tag can be included within the search result set. may not be shown. Thus, by including the appropriate tag (index or no index), it can be specified whether a page can be returned within a search result set.

브라우저와 같은 브라우저), 클라이언트 애플리케이션(들)(114) 및 VIP 로케이터 애플리케이션(116)을 도시한다.1 is an exemplary embodiment of a high-level client-server based network architecture 100 . Networked system 102, in the exemplary form of a network-based posting or payment system, provides server-side functionality to one or more client devices 110 over a network 104 (eg, the Internet or a wide area network (WAN)). . 1 illustrates, for example, a web client 112 running on a client device 110 (e.g., Internet Explorer® developed by Microsoft Corporation of Redmond, Wash.).

browser, such as a browser), client application(s) 114 and VIP locator application 116.

The client device 110 may be a wearable device, mobile phone, desktop computer, laptop, portable digital assistant (PDA), smart phone, tablet, ultrabook, netbook, laptop, multi-processor system, microprocessor-based or programmable consumer electronic device, It may be a game console, set top box, or any other communication device that a user may use to access networked system 102 . In some embodiments, the client device 110 includes a display module (not shown) that displays information (eg, in the form of a user interface). In another embodiment, the client device 110 may include one or more of a touch screen, an accelerometer, a gyroscope, a camera, a microphone, a Global Positioning System (GPS) device, and the like. Client device 110 may be a user's device used to conduct transactions related to digital items within networked system 102 . In one embodiment, the networked system 102 responds to requests for product listings, posts posts containing item listings of products available in the network-based posting system, and manages payments for these transactions. It is a system. One or more portions of network 104 may be an ad hoc network, an intranet, an extranet, a virtual private network (VPN), a local area network (LAN), a wireless LAN (WLAN), a wide area network (WAN), a wireless WAN (WWAN) ), Metropolitan Area Network (MAN), part of the Internet, part of the Public Switched Telephone Network (PSTN), cellular telephone network, wireless network, WiFi network, WiMax network, any other type of network, or any combination of two or more such networks. can do.

The client device 110 includes one or more client applications 114 (also referred to as " Also referred to as "App"). In some embodiments, when an e-commerce site application is included on client device 110, client application(s) 114 are configured to locally provide a user interface and functionality of client application(s) 114. At least a portion of is configured to communicate with the networked system 102 as needed for data or processing capabilities not available locally (eg, access to a database of salable items, user authentication, payment method verification). Conversely, if the e-commerce site application is not included on the client device 110, the client device 110 can use its web browser to access the e-commerce site (or a variant thereof) hosted on the networked system 102. can

One or more users 106 may be people, machines, or other means that interact with the client device 110 . In an exemplary embodiment, user 106 is not part of network architecture 100, but may interact with network architecture 100 via client device 110 or other means. For example, user 106 may provide input (eg, touch screen input or alphanumeric input) to client device 110 , and the input may be communicated over network 104 to networked system 102 . . In this example, networked system 102 may, in response to receiving input from user 106 , communicate information to be presented to user 106 to client device 110 over network 104 . In this way, user 106 may use client device 110 to interact with networked system 102 .

An application program interface (API) server 120 and web server 122 are connected to one or more application servers 140 and provide a programmatic interface and a web interface to the one or more application servers 140, respectively. Application server(s) 140 may host one or more publishing systems 142 and payment systems 144, each of publishing systems 142 and payment systems 144 may include one or more modules or applications. and may be implemented as hardware, software, firmware, or any combination thereof. Next, the application server(s) 140 are shown coupled to one or more information storage devices or database server(s) 124 enabling access to the database(s) 126. In an exemplary embodiment, database(s) 126 is a storage device that stores information (eg, posts or listings) to be posted to posting system 142 . Database(s) 126 may also store digital item information according to example embodiments.

Additionally, third party applications 132 running on third party server(s) 130 have programmatic access to networked system 102 via a programmatic interface provided by API server 120. is shown as For example, third party application 132 uses information retrieved from networked system 102 to support one or more features or functionality on a website hosted by the third party. The third party website provides, for example, one or more promotional functions, publishing functions, marketplace functions, or payment functions supported by associated applications on the networked system 102 .

Publishing system(s) 142 provides a number of publishing functions and services to users 106 accessing networked system 102 . Payment system(s) 144 likewise provide a number of functions for making or enabling payments and transactions. Although posting system(s) 142 and payment system(s) 144 are both shown in FIG. 1 as forming part of networked system 102, in alternative embodiments each system 142 and 144 may It will be appreciated that it may form part of a separate and distinct payment service from system 102 . In some embodiments, payment system(s) 144 may form part of posting system(s) 142 .

Selective indexing system 150 is operative to calculate a probability of future traffic for a given product and, based on this probability, select a set of view item pages corresponding to products with the highest traffic potential. provides a function to For example, selective indexing system 150 accesses a set of items, creates an ML model, and based on the ML model predicts a probability of future traffic for an item in the set of items. In some demonstrative embodiments, optional indexing system 150 is part of posting system(s) 142 .

Further, although the client-server based network architecture 100 shown in FIG. 1 uses a client-server architecture, the present invention is of course not limited to such an architecture, for example a distributed architecture system or a peer-to-peer architecture. It is easy to see that equally applicable in systems. The various posting system(s) 142, payment system(s) 144, and optional indexing system 150 may also be implemented as stand-alone software programs that do not necessarily have network functionality.

Web client 112 can access various posting and payment systems 142 and 144 through a web interface supported by web server 122 . Similarly, VIP locator application 116 accesses various services and functions provided by post and payment systems 142 and 144 through a programmatic interface provided by API server 120 . The VIP locator application 116, for example, allows a user to predict the probability of future traffic for a particular item in a set of items, and to display the item with the highest probability of future traffic for display on a display site in an offline manner via the networked system 102. Allows selection of the view item page corresponding to the item, and also creates an ML model to perform batch-mode communication between the VIP locator application 116 and the network system 102 .

2 is a block diagram illustrating the components of selective indexing system 150 that make up selective indexing system 150 to identify useful view item pages for selective indexing, according to some demonstrative embodiments. Optional indexing system 150 is shown as including data collection module 202, modeling module 204, indexing module 206, and item page selection module 208, all of which (e.g. , via a server, shared memory, or switch) to communicate with each other. Any one or more of these modules 202-208 may be implemented using one or more processors 210 (eg, by configuring such one or more processors 210 to perform the functions described for that module). and may thus include one or more of the processors 210 .

Any one or more of the described modules 202-208 may be implemented using only hardware (eg, one or more of the processors 210 of a machine), or using a combination of hardware and software. For example, any described module of selective indexing system 150 may include one or more of processors 210 configured to perform the operations described for that module (e.g., a subset of one or more processors of a machine). Alignment can be physically included. As another example, any module of selective indexing system 150 configures an alignment of one or more processors 210 (eg, among one or more processors of a machine) to perform the operations described for that module herein. may include software, hardware, or both. Accordingly, different modules of the selective indexing system 150 may contain and constitute a single arrangement of such processors 210 or different arrangements of such processors 210 at different points in time. Additionally, any two or more modules of selective indexing system 150 may be combined into a single module, and the functionality described herein for a single module may be subdivided among multiple modules. Further, according to various exemplary embodiments, modules described herein as being implemented within a single machine, database or device may be distributed across multiple machines, databases or devices.

3 is a flow diagram illustrating operation of the selective indexing system 150 when performing the method 300 of identifying useful view item pages based on indexed items, according to some demonstrative embodiments. As shown in FIG. 3 , one or more operations 302, 304, 306, 308, and 310 may be performed as part (e.g., a preemptive task, subroutine, or part) of method 300 according to some example embodiments.

At operation 302, data collection module 202 of optional indexing system 150 accesses a database (e.g., database(s) 126) to collect input variables associated with a product (or set of products). . Input variables include item level arguments. For example, collected item-level factors include the product's organic search traffic (i.e., how often the product is returned as an organic search result, as opposed to being presented as an advertisement), the product's view count (i.e., listing of the product), and so on. number of "clicks"), price, number of listings of a product (within a network-based publishing system), bounce count, number of independent sellers selling a product, quantity of product sold, length of time a product has been sold on the site; and search engine optimization factors such as keywords in the title of the post containing the listing of the product.

In some demonstrative embodiments, item level factors may be located in a local database (eg, database(s) 126 ) or third party server 130 . For example, data collection module 202 can retrieve item level arguments as raw, unedited data in a dump (eg, database dump).

At operation 304, the modeling module 204 uses a gradient boosted machine (GBM) to compute a probability of future traffic to the product, based on the input variables, a machine learning (ML) model. create data object Gradient boosting is a machine learning technique for regression and classification problems, which creates a predictive model in the form of an ensemble of weak predictive models (eg, a decision tree). Gradient boosting interactively combines weak learners into one strong learner. In some demonstrative embodiments, modeling module 204 may additionally use a product's previous SEO performance when calculating probabilities.

At operation 306 , the modeling module 204 trains the ML model based on the input variables collected by the collection module 202 . Supervised learning techniques can be applied to train ML models using input variables. In supervised learning scenarios, exemplary inputs (eg, input variables) and their desired outputs (eg, target variables) are provided to a model for the purpose of mapping inputs to desired outputs. The aggregation module 202 may use a sample set of products (either within the database 126 or on the third party server 130) over a predetermined period of time (e.g., 3 months) to identify search traffic values of products in the sample set. You can access and monitor the products in the set. The products in the sample set may include some of the products with zero search traffic values (no traffic at all), and some with greater than zero search traffic values (any amount of traffic). A target variable can be determined based on the search traffic values of the products in the sample set and applied to the ML model.

The modeling module 204 provides the ML model with the input variables and the product's organic search traffic as target variables. Based on the input and target variables, the modeling module 204 trains the ML model to maximize the probability of driving an organic search to a linked site by identifying (within the input variables) an item-level factor that corresponds to an increase in traffic. do. For example, modeling module 204 can identify a set of item level factors that correspond to a net increase in organic search traffic.

At operation 308, the indexing module 206 assigns an index state (eg, "index" or "no index") to a product (of the product set) based on the ML model. For example, the indexing module 206 may index a product as "indexed" or "no index" based on the probability of organic search traffic calculated by the ML model. In some embodiments, the indexing module 206 receives a threshold organic search traffic value and indexes the product based on the corresponding natural search traffic value and the item level factor of the product. For example, the threshold organic search traffic value may be based on the organic search traffic values of the products in the sample set monitored by the aggregation module 202 . In another exemplary embodiment, the indexing module 206 may receive a value representative of organic search traffic for a product set, and apply the product set's organic search traffic as a threshold.

At operation 310, the item page selection module 208 selects a useful item page (VIP) based on the indexed product. The VIP may correspond to the product with the highest probability of organic search traffic (based on the organic search traffic value probability calculated based on the ML model).

At operation 312, the item page selection module 208 assigns an "index" HTML tag to the VIP. For example, the item page selection module may simply tag the VIP with an "index" HTML tag, where the "index" HTML tag causes the VIP to be displayed in the middle of a search result set by a search engine. Pages not identified by the item page selection module 208 are tagged with the "no index" HTML tag to prevent search engines from retrieving the page for search requests.

Figure 4 shows how the execution of a machine learning model depends on a decision threshold (eg, threshold value). As shown in FIG. 4 , the x-axis 402 corresponds to the threshold value calculated by the modeling module 204 and the y-axis 404 corresponds to the percentage (eg, of gain or loss). As indicated by the graph 400 , the percentage along the y-axis 404 increases as the threshold value increases.

In some embodiments, to verify the performance of the ML model, the system creates a validation data set. The validation data set is generated by unbiased, random sampling of products that are distinct from (and not part of) the products in the sample set. To measure the performance of the ML model, validation metrics including decision thresholds and correlation between predicted and actual traffic are considered.

For a given decision threshold, the following values will be retrieved from the validation dataset: percentage of total VIP reduction, percentage of traffic loss due to that reduction, percentage of false negatives, percentage of purchased item loss, and percentage of GBM loss. can

5 is a diagram illustrating a process flow of a method 500 of identifying and indexing useful view item pages, in accordance with some demonstrative embodiments. Method 500 of FIG. 5 is described with reference to the interaction between selective indexing system 150 and database 126 . The database 126 may store a corpus of view item pages.

Modules, Components and Logic

Certain embodiments are described herein as comprising logic or a number of components, modules, or mechanisms. A module may consist of a software module (eg, code embedded on a machine readable medium or within a transmission signal) or a hardware module. A hardware module is a tangible unit that can perform specific operations and can be configured and arranged in specific ways. In an exemplary embodiment, one or more computer systems (eg, stand-alone client or server computer systems) or one or more hardware modules (eg, a processor or group of processors) of a computer system are operative to perform the specific operations described herein. It may be configured by software (eg, an application or application part) as a hardware module that does.

In various embodiments, hardware modules may be implemented mechanically or electronically. For example, a hardware module may include dedicated circuitry or logic (eg, a special purpose processor such as a Field-Programmable Gate Array (FPGA) or Application Specific Integrated Circuit (ASIC)) that is permanently configured to perform certain operations. there is. A hardware module may also include programmable logic or circuitry (eg, included within a general purpose processor or other programmable processor) that is temporarily configured by software to perform particular operations. The decision whether to implement a hardware module mechanically in dedicated, permanently configured circuitry or temporarily configured circuitry (eg, configured by software) may be made with cost and time considerations in mind.

Accordingly, the term “hardware module” refers to physically configured, permanently configured (e.g., hardwired) or temporarily configured (e.g., programmed) tangible entities. Considering embodiments in which hardware modules are temporarily configured (eg, programmed), each hardware module need not be configured or specified at any one time instance. For example, if a hardware module includes a general-purpose processor configured using software, the general-purpose processor may be configured with different special-purpose processors at different times. Thus, software, for example, configures a processor to implement a particular hardware module at one time instance and another hardware module at another time instance.

Hardware modules can provide information to and receive information from other hardware modules. Accordingly, the described hardware modules may be considered to be communicatively coupled. If multiple such hardware modules are present at the same time, communication may be achieved through signal transmissions connecting the hardware modules (eg, over an appropriate line or bus). In embodiments where multiple hardware modules are configured or instantiated at different times, communication between such hardware modules is accomplished, for example, by storing and retrieving information in memory structures accessed by the multiple hardware modules. For example, one hardware module may perform an operation and store the output of the operation in a communicatively connected memory device. Additional hardware modules can then access the memory device to retrieve and process the stored output at a later point in time. A hardware module may also initiate communication with an input or output device, and may operate on a resource (eg, a collection of information).

The various operations of example methods described herein may be performed at least in part by one or more processors that are either temporarily configured (eg, by software) or permanently configured to perform the associated operations. Whether temporarily or permanently configured, such processors may constitute processor-implemented modules that operate to perform one or more operations or functions. Modules referred to herein include processor implemented modules in some example embodiments.

Similarly, the methods described herein may be at least partially processor implemented. For example, at least some of the operations of a method may be performed by one or more processors or processor-implemented modules. The performance of a particular operation may be distributed among one or more processors, both residing within a single machine and distributed across multiple machines. In some demonstrative embodiments, the processor or processors may be located within a single location (eg, within a home environment, office environment, or as a server farm), while in other embodiments, a processor may be distributed across multiple locations. It can be.

In addition, one or more processors may operate to support performance of related operations in a “cloud computing” environment or as “software as a service” (SaaS). For example, at least some of the operations may be performed by a group of computers (as an example of a machine including a processor), such operations may be performed by a network (eg, the Internet) and one or more appropriate interfaces (eg, an application program interface (eg, an application program interface)). It is accessible through the API)).

electronic devices and systems

Exemplary embodiments may be implemented in digital electronic circuitry, in computer hardware, firmware, software, or combinations thereof. An exemplary embodiment may be a computer program product, eg, within an information carrier, such as a machine readable medium, for execution by or for controlling the operation of a data processing device, such as a programmable processor, computer or multiple computers. It may be implemented using a specifically recorded computer program.

A computer program may be written in any form of programming language, including a compiled or interpreted language, and may be written in any form, for example as a stand-alone program suitable for use in a computer environment or as modules, subroutines or other units. can be placed as A computer program can be deployed to be executed on one computer or on multiple computer devices at one site or distributed across multiple sites and interconnected by a communication network.

In an example embodiment, operations may be performed by one or more programmable processors executing functions by executing a computer program to process input data and generate output. Operations of the method may also be performed by the apparatus of the exemplary embodiment, which apparatus may be implemented as a special purpose logic circuit such as a field programmable gate array (FPGA) or an application specific integrated circuit (ASIC).

A computing system may include a client and a server. A client and server are generally remote from each other and typically interact through a communication network. The relationship of client and server is formed by computer programs running on the respective computers and having a client-server relationship to each other. It will be appreciated that in embodiments deploying a programmable computing system, both hardware architecture and software architecture must be considered. Specifically, whether a particular function is implemented in permanently configured hardware (e.g., an ASIC), temporarily configured hardware (e.g., a combination of software and a programmable processor), or in permanently configured hardware and temporarily. The choice of implementing a combination of configured hardware is a design choice. The following describes hardware (eg, machine) and software architectures that may be deployed in various example embodiments.

Exemplary Machine Architecture and Machine-readable Media

5 is a component of a machine 500 capable of reading instructions from a machine readable medium (eg, a machine readable storage medium) and performing any one or more methods discussed herein, according to some demonstrative embodiments. It is a block diagram showing Specifically, FIG. 5 illustrates the execution of instructions 516 (eg, software, programs, applications, applets, apps, or other executable code) that cause machine 500 to perform any one or more methods discussed herein. A graphical representation of machine 500 is shown in an exemplary form of a computer system that can be used. Additionally or alternatively, the instructions may implement the modules of FIG. 2 . The instructions transform a general-purpose, unprogrammed machine into a specific machine programmed to perform the functions described and illustrated in the manner described. In an alternative embodiment, machine 500 may operate as a stand-alone device or may be connected (networked) to other machines. In a networked deployment, machine 500 can operate as a server machine or a client machine in a server-client network environment, or as a peer machine in a peer-to-peer (or distributed) network environment.

The machine 500 is a server computer, a client computer, a personal computer (PC), a tablet computer, a laptop computer, a netbook, a set-top box (STB), a personal digital assistant (PDA), an entertainment media system, a cellular phone, a smart phone, and a mobile device. , a wearable device (eg, smart watch), a smart home device (smart appliance), another smart device, a web appliance, a network router, a network switch, a network bridge, or an action to be taken by the machine 500 any machine capable of executing the specified instructions 516 sequentially or otherwise. Further, although only a single machine 500 is shown, the term “machine” also refers to a collection of machines 500 that individually or jointly execute instructions 516 to perform any one or more methods described herein. should be treated as including

Machine 500 may include a processor 510 , memory 530 and I/O components 550 that may be configured to communicate with each other via, for example, a bus 502 . In an exemplary embodiment, the processor 510 (eg, a central processing unit (CPU), a reduced instruction set computing (RISC) processor, a complex instruction set computing (CISC) processor, a graphics processing unit (GPU), a digital signal processor (DSP) ), an application specific integrated circuit (ASIC), a radio-frequency integrated circuit (RFIC), another processor, or any suitable combination thereof) may include, for example, a processor 512 capable of executing instructions 516 and a processor ( 514) may be included. The term "processor" is intended to include multi-core processors, which may include two or more independent processors (sometimes referred to as "cores") capable of executing instructions concurrently. 5 illustrates multiple processors, machine 500 may include a single processor with a single core, a single processor with multiple cores (eg, a multi-core processor), multiple processors with a single core, multiple cores. It may include multiple processors having, or any combination thereof.

Memory/storage 530 may include memory 532 and storage unit 536, such as main memory or other memory storage, both accessible to processor 510 via bus 502, for example. do. The storage unit 536 and memory 532 store instructions 516 that implement any one or more methods or functions described herein. Instructions 516 may also, during execution by machine 500, reside within memory 532, within storage unit 536, within at least one processor 510 (e.g., within a cache memory of a processor), or any of these. It may reside fully or partially in any suitable combination. Thus, memory 532, storage unit 536, and memory of processor 510 are examples of machine-readable media.

As used herein, "machine-readable medium" means a device capable of temporarily or permanently storing instructions or data, and includes random-access memory (RAM), read-only memory (ROM), buffer memory , flash memory, optical media, magnetic media, cache memory, other types of storage devices (eg EEPROM (Erasable Programmable Read-Only Memory)), and/or any suitable combination thereof. . The term “machine-readable medium” should be taken to include a single medium or multiple mediums (eg, centralized or distributed databases, or associated caches and servers) that can store instructions 516 . The term "machine-readable medium" is also taken to include any medium or combination of multiple media that can store instructions (eg, instructions 516) to be executed by a machine (eg, machine 500). The instructions, when executed by one or more processors (e.g., processor 510) of machine 500, thereby cause machine 500 to perform any one or more methods described herein. do. Accordingly, a "machine-readable medium" refers to a single storage device or device as well as a "cloud-based" storage system or storage network that includes multiple storage devices or devices.

I/O component 550 may include various components for receiving input, providing output, generating output, transferring information, exchanging information, and capturing measurements. The particular I/O components 550 included in a particular machine will depend on the type of machine. For example, a portable machine such as a mobile phone will include a touch input device or other input mechanism, while a headless server machine will probably not include such a touch input device. It will be appreciated that I/O component 550 may include many other components not shown in FIG. 5 . I/O components 550 are grouped according to function only to simplify the following discussion, and this grouping is not limiting. In various demonstrative embodiments, I/O component 550 may include output component 552 and input component 554 . The output component 552 may include a visual component (e.g., a plasma display panel (PDP), a light emitting diode (LED) display, a liquid crystal display (LCD), a projector, or a cathode ray tube (CRT)), an acoustic component (e.g., a speaker), haptic components (eg, vibration motors, resistive mechanisms), other signal generators, and the like. Input component 554 may be an alphanumeric input component (eg, a keyboard, a touch screen configured to receive alphanumeric input, a photo-optical keyboard, or other alphanumeric input component), a point-based input component (eg, For example, a mouse, touchpad, trackball, joystick, motion sensor, or other pointing device), tactile input component (e.g., a physical button, a touch screen that provides touch or position and/or strength of a touch gesture, or other tactile input) component), an audio input component (eg, a microphone), and the like.

In other illustrative embodiments, I/O component 550 may include biometric component 556, motion component 558, environment component 560, or location component 562, among various other components. For example, the biometric component 556 detects expressions (eg, hand gestures, facial expressions, voice expressions, gestures, or eye tracking), and detects biosignals (eg, blood pressure, heart rate, body temperature, sweat, or brain waves). ) and identify a person (eg, voice identification, retinal identification, face identification, fingerprint identification, or electroencephalogram-based identification). Motion component 558 can include an acceleration sensor component (eg, accelerometer), a gravity sensor component, a rotation sensor component (eg, gyroscope), and the like. Environment component 560 may include, for example, a light sensor component (eg, a photometer), a temperature sensor component (eg, one or more thermometers that detect ambient temperature), a humidity sensor component, a pressure sensor component (eg, a barometer). , acoustic sensor components (eg, one or more microphones to detect ambient noise), proximity sensor components (eg, infrared sensors to detect nearby objects), gas sensors (eg, to detect concentrations of harmful gases for safety or pollution of the atmosphere). gas detection sensors that measure substances), or other components that can provide indications, measurements, or signals corresponding to the surrounding physical environment. Position component 562 may include a position sensor component (e.g., a Global Position System (GPS) receiver component), an altitude sensor component (e.g., an altimeter or barometer that detects air pressure from which altitude may be derived), a direction sensor component (e.g., , magnetometer) and the like.

컴포넌트(예컨대, Bluetooth￠

저에너지), Wi-Fi￠

컴포넌트, 및 다른 방식으로 통신을 제공하는 기타 통신 컴포넌트를 포함할 수 있다. 장치(570)는 또 다른 머신 또는 임의의 다양한 주변 장치(예를 들어, USB(Universal Serial Bus)를 통해 연결된 주변 장치)일 수 있다. Communication can be implemented using a wide variety of technologies. I/O component 550 may include a communication component 564 operable to connect machine 500 to network 580 or device 570 via connection 582 or connection 572, respectively. For example, communication component 564 may include a network interface component or other suitable device for interfacing with network 580 . In another example, communication component 564 may be a wired communication component, a wireless communication component, a cellular communication component, a near field communication (NFC) component, a Bluetooth®

component (e.g. Bluetooth

low energy), Wi-Fi￠

component, and other communication components that provide communication in other ways. Device 570 may be another machine or any of a variety of peripherals (eg, peripherals connected via a Universal Serial Bus (USB)).

신호 삼각측량을 통해 얻어지는 위치, 특정 위치를 표시하는 NFC 비컨 신호의 검출을 통해 얻어지는 위치 등과 같은 다양한 정보가 통신 컴포넌트(564)를 통해 도출될 수 있다. Additionally, communication component 564 can detect an identifier or include a component capable of detecting an identifier. For example, the communications component 564 may include a radio frequency identification (RFID) tag reader component, an NFC smart tag detection component, an optical reader component (e.g., a one-dimensional barcode such as a Universal Product Code (UPC) barcode, a Quick Response (QR) barcode). ) code, Aztec code, Data Matrix, Dataglyph, MaxiCode, PDF417, Ultra code, multi-dimensional barcodes such as UCC RSS-2D barcodes, and other optical codes), or acoustic detection components (e.g., microphones to identify tagged audio signals) can include In addition, location obtained through Internet Protocol (IP) geo-location, Wi-Fi

A variety of information may be derived via communication component 564, such as location obtained through signal triangulation, location obtained through detection of an NFC beacon signal indicative of a particular location, and the like.

transmission medium

네트워크, 다른 유형의 네트워크, 또는 둘 이상의 이러한 네트워크의 조합일 수 있다. 예를 들어, 네트워크(580) 또는 네트워크(580)의 일부는 무선 또는 셀룰러 네트워크를 포함할 수 있고, 연결부(582)는 CDMA(Code Division Multiple Access) 접속, GSM(Global System for Mobile communications) 접속, 또는 다른 유형의 셀룰러 또는 무선 연결을 포함할 수 있다. 본 예시에서, 연결부(582)는 단일 캐리어 무선 전송 기술(1xRTT), EVDO(Evolution-Data Optimized) 기술, 일반 패킷 무선 서비스(GPRS) 기술, EDGE(Enhanced Data rates for GSM Evolution) 기술, 3G를 포함하는 3GPP(third Generation Partnership Project), 4 세대 무선(4G) 네트워크, UMTS(Universal Mobile Telecommunications System), HSPA(High Speed Packet Access), WiMAX(Worldwide Interoperability for Microwave Access), LTE(Long Term Evolution) 표준, 다양한 표준 설정 기관에 의해 정의된 다른 기술, 다른 장거리 프로토콜 또는 다른 데이터 전송 기술과 같은 임의의 다양한 유형의 데이터 전송 기술을 구현할 수 있다.In various exemplary embodiments, one or more portions of network 580 may be an ad hoc network, an intranet, an extranet, a virtual private network (VPN), a local area network (LAN), a wireless LAN (WLAN), a wide area network (WAN), a wireless WAN. (WWAN), metropolitan area network (MAN), Internet, part of the Internet, part of the Public Switched Telephone Network (PSTN), plain old telephone service (POTS) network, cellular telephone network, wireless network, Wi-Fi￠

It may be a network, another type of network, or a combination of two or more such networks. For example, the network 580 or part of the network 580 may include a wireless or cellular network, and the connection unit 582 may include a Code Division Multiple Access (CDMA) connection, a Global System for Mobile communications (GSM) connection, or other types of cellular or wireless connections. In this example, the connection unit 582 includes single carrier radio transmission technology (1xRTT), Evolution-Data Optimized (EVDO) technology, General Packet Radio Service (GPRS) technology, Enhanced Data rates for GSM Evolution (EDGE) technology, and 3G Third Generation Partnership Project (3GPP), fourth generation wireless (4G) networks, Universal Mobile Telecommunications System (UMTS), High Speed Packet Access (HSPA), Worldwide Interoperability for Microwave Access (WiMAX), Long Term Evolution (LTE) standards, Any of the various types of data transmission technologies may be implemented, such as other technologies defined by various standard-setting bodies, other long-range protocols, or other data transmission technologies.

Instructions 516 use a transmission medium over a network interface device (e.g., a network interface component included in communication component 564) and use any of a number of well-known transport protocols (Hyper Text Transfer Protocol (HTTP)). may be transmitted or received over network 580 using either. Similarly, instructions 516 may be transmitted or received using a transmission medium over connection 572 to device 570 (eg, a peer-to-peer connection). The term "transmission medium" includes a digital or analog communication signal or other intangible medium capable of storing, encoding, or carrying instructions 516 for execution by machine 500 and enabling communication of such software. It should be treated as including any intangible medium. A transmission medium is one embodiment of a machine readable medium.

Examples numbered below are examples.

1. As a system,

one or more processors;

A memory storing instructions that, when executed by at least one of the one or more processors, cause the system to perform operations;

the actions

collecting input variables from the database, which input variables include item level factors of the product;

generating a machine learning model data object based on the input variable;

assigning a target variable to a machine learning model data object;

training a machine learned model data object based on a target variable and an input variable;

Calculating a natural search traffic probability of a product based on the trained machine learning model data object;

and indexing the product based on the item level factor of the product and the organic search traffic probability.

2. The system of embodiment 1, wherein the machine learning model data object is a gradient boosted machine model.

3. The system of embodiment 1 or embodiment 2, wherein the item level factor is

the product's prior natural search traffic; and

a view count of the product;

a watch count of the product;

a bounce count of the product and

Include the quantity of products sold.

4. The system of any one of embodiments 1-3, wherein the target variable includes organic search traffic values.

5. The system of any one of embodiments 1-4, wherein the instructions cause the system to perform operations to identify useful item pages corresponding to the product based on the product's item level factor and natural search traffic probability.

6. The system of any one of embodiments 1 to 5, wherein the instructions for indexing the product based on the item level factor of the product and the organic search traffic probability include indexing status based on the item level factor of the product and the organic search traffic probability. This includes assigning them to products.

7. The system of any one of embodiments 1-6, wherein the instructions to cause training of the machine learning model data object based on the target variable and the input variable comprises correlating a set of input variables to the target variable. do.

8. As a computer-implemented method,

collecting input variables from the database, the input variables including item level parameters of the product;

generating a machine learning model data object based on the input variables;

assigning target variables to machine learning model data objects;

training a machine learning model data object based on a target variable and an input variable;

Calculating a probability of natural search traffic for a product based on the trained machine learning model data object;

Indexing the product based on the product's item level factor and organic search traffic probability.

9. The method of embodiment 8, wherein the machine learning model data object is a gradient boosted machine model.

10. The method of embodiment 8 or 9, wherein the item level factor is

Your product’s previous organic search traffic;

the view count of the product;

the watch count of the product;

the bounce count of the product;

Include the quantity of products sold.

11. The method of any one of embodiments 8 to 10, wherein the target variable includes an organic search traffic value.

12. The method of any one of embodiments 8 to 11, further comprising identifying useful item pages corresponding to the product based on the item level factor of the product and the organic search traffic probability.

13. The method of any one of Embodiments 8 to 12, wherein indexing the product based on the item-level factor of the product and the probability of natural search traffic includes determining an index state based on the item-level factor of the product and the probability of organic search traffic. Allocating to

14. The method of any of embodiments 8-13 wherein training the machine learning model data object based on the target variable and the input variable comprises correlating a set of input variables to the target variable. .

15. A machine readable medium carrying instructions that, when executed by one or more processors of a machine, cause a machine to perform operations;

the actions

collecting input variables from the database, which input variables include item level parameters of the product;

generating a machine learning model data object based on the input variable;

assigning a target variable to a machine learning model data object;

\*106 training a machine learning model data object based on a target variable and an input variable;

Calculating a natural search traffic probability of a product based on the trained machine learning model data object;

and indexing the product based on the item level factor of the product and the organic search traffic probability.

16. The machine readable medium of embodiment 1, wherein the machine learning model data object is a gradient boosted machine model.

17. The machine readable medium of embodiment 15 or embodiment 16, wherein the item level factor is

Your product’s previous organic search traffic;

the view count of the product;

\*113 product's watch count,

the bounce count of the product;

Include the quantity of products sold.

18. The machine readable medium of any one of embodiments 15-17, wherein the target variable comprises an organic search traffic value.

19. The machine readable medium of any one of embodiments 15-18, wherein the instructions cause the system to perform operations to identify useful item pages corresponding to the product based on the product's item level factor and natural search traffic probabilities. .

20. The machine readable medium of any one of embodiments 15-19, wherein the instructions to index the product based on the product's item level factor and the organic search traffic probability cause the system to: An operation of assigning an index state to a product based on is performed.

language

Multiple instances may implement a component, operation, or structure described herein as a single instance. Although individual operations of one or more methods are shown and described as separate operations, one or more individual operations may be performed concurrently, and the operations need not be performed in the order shown. Structures and functionality presented as separate components in example configurations may be implemented as a combined structure or component. Similarly, structures and functionality presented as a single component may be implemented as separate components. These and other variations, modifications, additions and improvements are within the scope of the subject matter herein.

Although the summary of the present invention has been described with reference to specific exemplary embodiments, various modifications and changes may be made to these embodiments without departing from the broader scope of the present invention. Such embodiments of the invention herein, where in fact two or more are disclosed, are not intended to voluntarily limit the scope of this application to any single disclosure or invention, but are individually or collectively referred to as an "invention" merely for convenience. term can be referred to.

The embodiments described herein are described in sufficient detail to enable any person skilled in the art to practice the disclosed teachings. Other embodiments may be used and derived therefrom, so that structural and logical substitutions and changes may be made without departing from the scope of the present disclosure. Therefore, the detailed description is not to be taken in a limiting sense, and the scope of the various embodiments is defined only by the appended claims and the full scope of equivalents to which such claims are entitled.

As used herein, the term “or” may be interpreted in an inclusive or exclusive sense. Also, multiple instances may be provided for a resource, operation, or structure described herein as a single instance. Further, the boundaries between the various resources, operations, modules, engines, and data stores are somewhat arbitrary, and specific operations are described in relation to specific example configurations. Other assignments of functionality are envisioned and may be within the scope of various embodiments of the present disclosure. In general, structures and functionality presented as separate resources in the example configurations may be implemented as a combined structure or resource. Similarly, structures and functionality presented as a single resource may be implemented as separate resources. These and other variations, modifications, additions and improvements are within the scope of the embodiments of the present disclosure as indicated by the appended claims. Accordingly, the specification and drawings are to be regarded in an illustrative rather than restrictive sense.

Claims (15)

As a system,
one or more processors;
A memory storing instructions that cause the system to perform operations when executed by at least one of the one or more processors;
These actions are
Collecting input variables from the database, the input variables including item level factors of a product;
generating a machine learned model data object based on the input variable;
assigning a target variable to the machine learning model data object;
training the machine learning model data object based on the target variable and the input variable;
calculating a natural search traffic probability of the product based on the trained machine learning model data object;
Indexing the product based on the item level factor of the product and the natural search traffic probability
system.
According to claim 1,
The machine learning model data object is a gradient boosted machine model
system.
According to claim 1,
The item level argument is
a prior natural search traffic of the product;
a view count of the product;
a watch count of the product;
a bounce count of the product;
including the quantity of the product sold
system. According to claim 1,
The target variable includes an organic search traffic value
system.
According to claim 1,
The instructions further cause the system to identify a valuable item page corresponding to the product based on the item level factor of the product and the organic search traffic probability.
system.
According to claim 1,
Indexing the product based on the item level factor of the product and the organic search traffic probability includes assigning an index status to the product based on the item level factor of the product and the organic search traffic probability. doing
system. According to claim 1,
Training the machine learning model data object based on the target variable and the input variable comprises correlating a set of input variables among the input variables to the target variable.
system. As a computer implemented method,
collecting input variables from the database, the input variables including item level parameters of the product;
generating a machine learning model data object based on the input variable;
assigning a target variable to the machine learning model data object;
training the machine learning model data object based on the target variable and the input variable;
Calculating a natural search traffic probability of the product based on the trained machine learning model data object;
Indexing the product based on the item level factor of the product and the organic search traffic probability
A computer implemented method. According to claim 8,
The machine learning model data object is a gradient boosted machine model
A computer implemented method. According to claim 8,
The item level argument is
previous organic search traffic for the product;
a view count of the product;
a watch count of the product;
a bounce count of the product;
including the quantity of the product sold
A computer implemented method. According to claim 8,
The target variable includes an organic search traffic value
A computer implemented method. According to claim 8,
The method further comprises identifying a useful item page corresponding to the product based on the item level factor of the product and the organic search traffic probability.
A computer implemented method. According to claim 8,
Indexing the product based on the item level factor of the product and the organic search traffic probability includes assigning an index status to the product based on the item level factor of the product and the organic search traffic probability. doing
A computer implemented method. According to claim 8,
Wherein training the machine learning model data object based on the target variable and the input variable comprises correlating a set of the input variables to the target variable.
A computer implemented method. A machine-readable storage medium storing instructions that, when executed by one or more processors of a machine, cause the machine to perform operations, comprising:
These actions are
collecting input variables from the database, the input variables including item level parameters of the product;
generating a machine learning model data object based on the input variable;
assigning a target variable to the machine learning model data object;
training the machine learning model data object based on the target variable and the input variable;
Calculating a natural search traffic probability of the product based on the trained machine learning model data object;
Indexing the product based on the item level factor of the product and the natural search traffic probability
A machine-readable storage medium.

Images